On Saturday, I spent the day at the Vintage Computer Federation’s museum in Wall, NJ working on the 1981 Heathkit Hero 1 in their collection. I have been designated not only the Hero expert of the VCFed, but the man suited for the job of repairing it. I brought a massive toolkit worth of stuff, most of which turned out to be unnecessary.
I removed the old batteries from 2005, then tried out my modern 6V 5Ah batteries, which work alright but are too wide to fit in the head compartment. The aluminum sheet metal battery retainer was built for square batteries, not rectangular ones. I have an idea of how to make a custom sized piece at my local hackerspace to avoid modifying the original part.
Seems that this robot has had a few internal modifications. It's running a home made EPROM with a demonstration program instead of the default -- I'll have to dump the contents sometime in the future to see what’s on there. On the Main Drive board, a relay has been removed and replaced with a modern professional custom made board from 2006 (with an surface mount chip). Whomever installed that also knew what they were doing, and must have had a good reason to make such a change.
The I/O board seems to be causing major faults for the Hero. Initially, the robot spouted nothing but gibberish and garbled text on its display, along with jerky unchoreographed movements. The backside of the board is covered in wirewrap, apparently to fix some traces that are not in great shape. An accompanying page of correspondence between Robert Doerr (the premier Hero robot expert) and the last owner confirms that the I/O board has issues. Apparently every time the I/O board flexes, the traces start losing continuity.
I also swapped in the spare I/O board that the last owner had bought, however it yielded only a silent and unresponsive robot. I swapped the first I/O board back in, and it started showing me useful information on the hexadecimal keypad and display, and some promise that not all was lost. I checked through some contents of the ROM via the keypad, and it responded correctly most of the time. When I polled devices that were accessed through the I/O board, it didn’t pass the diagnostic tests. It tries to speak when prompted to, but it fails to say anything other than AAAAAAAAAAAHHHHHHHH. At best, I was able to enter a few things on the keypad, and get back some correct results according to the diagnostic booklet. However, many critical tests are failing.
Two other VCFed members, Dean and his son Drew, cleaned up the plastic panels, so the at least the robot looks nice! You might remember their exhibit from VCF East XI.
While I was unsuccessful in repairing the Hero 1 in a day, I did learn a ton about it.
